Skills
skills/mukul975/anthropic-cybersecurity-skills/processing-stix-taxii-feeds/Gen Agent Trust Hub

processing-stix-taxii-feeds

Pass

Audited by Gen Agent Trust Hub on Apr 20, 2026

Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted threat intelligence data from external TAXII servers, which constitutes an indirect prompt injection surface. \n
  • Ingestion points: scripts/agent.py fetches STIX bundles from remote URLs provided by the user. \n
  • Boundary markers: Absent; the skill categorizes and prints intelligence data as plain text. \n
  • Capability inventory: Network read/write capabilities are used to interact with TAXII servers. \n
  • Sanitization: While the stix2 library is used for schema validation, it does not sanitize against malicious natural language instructions embedded in STIX fields. \n- [DATA_EXFILTRATION]: The scripts/agent.py script performs network operations to non-whitelisted, user-provided TAXII server endpoints.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 20, 2026, 05:05 AM